Download (BMI) in Turner syndrome before and during growth hormone (GH)

International Journal of Obesity (2000) 24, 232±235
ß 2000 Macmillan Publisher Ltd All rights reserved 0307±0565/00 $15.00
www.nature.com/ijo
Body mass index (BMI) in Turner syndrome
before and during growth hormone (GH)
therapy
PR Blackett1*, AC Rundle2, J Frane2 and SL Blethen2
1
Department of Pediatrics, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73190, USA; and 2Genentech Inc., South
San Francisco, CA 94080-4990, USA
OBJECTIVE: To study whether body mass index (BMI) is different in girls with Turner syndrome (TS) compared to
normal girls, and whether BMI in TS is affected by growth hormone (GH) treatment.
DESIGN: A retrospective cross-sectional study.
SUBJECTS: 2468 girls with TS enrolled in the National Cooperative Group Study (NCGS), a collaborative surveillance
study for assessing GH-treated children.
MEASUREMENTS: BMI and BMI standard deviation score (BMI SDS) at baseline and during GH treatment were
computed from height and weight data.
RESULTS: BMI in TS patients increases with age as expected. However, BMI SDS increased starting at about age 9 y. A
similar pattern of increase in BMI SDS was observed after each year of GH treatment for up to 4 y, but GH treatment
did not change the magnitude of increase. BMI and BMI SDS curves before and during GH treatment were essentially
superimposable.
CONCLUSION: These ®ndings suggest that mechanisms speci®c for TS are responsible for the age-related increase in
BMI SDS. This increase was unaffected by GH treatment.
International Journal of Obesity (2000) 24, 232±235
Keywords: Turner syndrome; body mass index; obesity; growth hormone
Introduction
Methods
Short stature is one of the most common phenotypic
features of Turner syndrome (TS).1 Growth hormone
(GH) alone or in combination with oxandrolone can
increase adult height in TS, and GH treatment is now
the standard of care for girls with TS in many
countries.1 ± 4 Girls with TS are thought to be prone
to gain excessive weight, which may be a risk factor
for developing diabetes and cardiovascular disease in
adult life.5 ± 7 For this reason, we were interested in
studying the effect of GH treatment on body mass
index (BMI) in order to determine if GH in¯uences
body composition in ways unrelated to linear growth.
We chose to study BMI because it has been validated
as an indicator of obesity in children and is widely
used for this purpose.8,9 Furthermore, height and
weight, from which the BMI could be computed,
were available to us.
Data from the National Cooperative Growth Study
(NCGS)1 were analyzed to determine whether there
would be changes in BMI in TS girls relative to
normal girls during GH treatment.
Description of the patients
*Correspondence: PR Blackett, Department of Pediatrics,
University of Oklahoma Health Sciences Center, PO Box 26901,
Oklahoma City, OK 73190, USA.
E-mail: [email protected]
Received 19 May 1999; revised 7 September 1999; accepted
13 September 1999
The NCGS was established in 1985 to monitor the
safety and effectiveness of Genentech GH products.
The methods of subject enrollment and data collection
have been described elsewhere.1 As of 30 June, 1998,
272 children had been enrolled in the NCGS database.
There were 3114 girls with TS. This analysis is
con®ned to the 2468 girls who had not been treated
with GH before enrolling in NCGS and who remained
prepubertal throughout the 4 y of observation. Height
measurements were performed at the participating
centers using a Harpenden stadiometer or other wallmounted device as speci®ed in the NCGS protocol.
Weight measurements were carried out using a balance beam. BMI was calculated as the weight in
kilograms divided by the height in meters squared.
BMI standard deviation score (SDS) was calculated
using data from the First National Health and Nutrition Examination Study (NHANES).11
Analysis
Statistical analyses and graphics were carried out
using `Splus' for programming.12 Both BMI SDS
and BMI values were plotted against age at enrollment
and age after the 1st, 2nd, 3rd and 4th year of GH
BMI in Turner syndrome
PR Blackett et al
233
treatment. Smoothed curves (lowess) were plotted for
the median BMI for age and the mean BMI SDS.
Results
When BMI at enrollment was plotted against age at
enrollment in TS (Figure 1A), there was an increase in
BMI with age similar to that observed in the normal
girls.11 Median BMI was unaffected by GH treatment
for a period of up to 4 y. This is indicated by the fact
that the curves for BMI before and during GH treatment are superimposable (Figure 1B). When BMI
SDS at enrollment was plotted against age, the mean
BMI SDS for younger girls was close to zero with
most values between plus and minus two standard
deviations, indicating that BMI is relatively normal in
young TS girls. At about age 9 ± 10 y, there was a
progressive increase in the mean BMI SDS (Figure
2A), indicating that TS girls have increased BMI
relative to unaffected girls at these ages. After age
12 y, the mean BMI SDS curve was relatively ¯at, but
the mean BMI SDS remained greater than the values
for younger girls. Similar age-related increases in
BMI SDS occurred after GH treatment. However,
the curves of BMI SDS vs age were almost superimposable, indicating no effect of GH on BMI SDS
(Figure 2B).
Discussion
Since percentiles and smoothed percentile curves have
become available both in the US and in the UK,
studies have supported the use of BMI as an index
of adiposity in children.8,9,13 ± 15 It has also been
validated using dual X-ray absorptiometry as the
standard16 and recommended for assessing obesity
by an expert panel.17 A high BMI has added signi®cance in childhood since it predicts adiposity in
adulthood.18 However the signi®cance of BMI ¯uctuations in children with altered phenotypes affecting
body proportions is unclear. This study provides
information on a large number of girls with TS and
a wide range of BMI values, showing that weight for
stature indices such as BMI may lack speci®city as an
index of body fat.19 Nevertheless analysis of a large
set of data provides evidence on whether there is a
tendency for girls with TS to become obese and serves
as a possible guide to further studies on body composition analysis using other methods.19
Our results con®rm previous observations that older
girls with TS are relatively overweight as a group.6
Cross-sectional analysis of the TS girls before GH
treatment showed that an increase in BMI SDS
occurred at age 9 ± 10 y. Because computation of the
BMI SDS was based on normal girls in NHANES I,
Figure 1 (A) BMI at enrollment is plotted against age for
individual girls with TS. The smoothed curve shows the
median BMI before GH treatment. (B) Smoothed curves of
median BMI vs age for Turner girls before GH (ÐÐÐ), and
after 1 ( ), 2 (- - -), 3 ( ± ± ± ), and 4 ( Ð Ð Ð ) y of GH treatment. Note that the curves are essentially superimposable.
International Journal of Obesity
BMI in Turner syndrome
PR Blackett et al
234
Figure 2 (A) BMI SDS at enrollment was plotted against age for
individual girls with TS. The smoothed curve shows the mean
BMI SDS. (B) Smoothed curves for mean BMI SDS vs age for
Turner girls before GH (ÐÐ), and after 1 ( ), 2 (- - -), 3 ( ± ± ± ),
and 4 ( Ð Ð Ð ) y of GH treatment. Note that the curves are
essentially superimposable.
International Journal of Obesity
the increase in BMI SDS may re¯ect an observation
speci®c for TS. A BMI SDS based on normal girls as
the standard provided a way to contrast girls with TS,
nevertheless the normal girls in the NHANES study
would have started puberty within a normal age range,
whereas TS girls did not. Since normal girls increase
their adiposity during pubertal development,20,21 the
fact that TS girls underwent a relatively excessive
gain in BMI represented by the SDS is all the more
signi®cant.
Since the change in BMI SDS with age was similar
after each of the ®rst 4 y of GH treatment, there was no
recognizable effect of GH on BMI SDS. Our results
are consistent with Corel et al,22 who reported that the
leaner TS girls treated with GH had an increase in BMI
SDS. Interestingly, this group also observed that those
with a BMI SDS greater than 2 at baseline had an
initial decrease consistent with an initial lipolytic
effect due to growth hormone. Also, the patients in
this multi-national study were mostly between ages 9
and 13 y when they began GH. This is the age range
where we found an increase in BMI SDS that occurred
both with and without GH treatment. Thus, we suggest
that the changes in BMI SDS reported in TS girls were
not due to GH treatment but merely the re¯ection of a
process that occurs in TS girls as they grow older. This
suggestion is supported by a study of heights and
weights in TS girls from the Netherlands, Denmark
and Sweden.23 In this study, TS girls showed a disproportionate increase in weight at adolescent ages.
Although the data were not analyzed by age, the shift
in weight for height percentiles was found for heights
greater than 120 cm. This is the approximate height of
9-y-old girls with TS. Thus our results con®rm and
complement those of Rongen-Westerlaken et al23 and
suggest a disproportionate increase in adiposity in TS
patients in the second decade of life. These data do not
support a role for GH in the process.
We propose three possible explanations for the agerelated increase in BMI SDS in TS. These could also
act in combination. First, the onset of adrenarche
without gonadal changes of puberty could result in
the absence of ovarian steroids to balance increasing
levels of adrenal androgens.24 This hormonal imbalance could then result in the observed changes in body
composition. Second, the increase in weight relative
to height could be a compensatory phenomenon by
which the girls with TS have a tendency to achieve a
`critical weight' required for the onset of menarche as
proposed by Frisch and McArthur.25 When menarche
fails to occur weight gain may continue past a set
point normally regulated by the onset of puberty.
Third, decreased limb relative to trunk growth results
in the short stocky physique and an increased weight
for height characteristic of girls with TS.26 The
appearance of these characteristics may be re¯ected
by the upward swing in the BMI SDS at age 9 y. A
relative increase in somatic tissue in TS is supported
by the ®nding that skin-fold thickness is less than in
children with similar weight for height.4
BMI in Turner syndrome
PR Blackett et al
The age-related increase in BMI SDS appears to be
unique for TS and is unaffected by GH treatment. This
®nding supports the development of more speci®c
studies to de®ne what changes in body composition
may occur and the mechanisms responsible.
References
1 Saenger P. Turner's syndrome. New Engl J Med 1996; 335:
1749 ± 1754.
2 Van den Broeck J, Massa GG, Attanasio A, Metranga A,
Chaussain JL, Price DA, Aarskog D, Wit JM. Final height
after long-term growth hormone treatment in Turner syndrome. J Pediatr 1995; 127: 729 ± 735.
3 Nilsson KO, Albertsson-Wikland K, Alm J, Aronson S,
Gustafsson J, Hagenas L, Hager A, Ivarsson SA, Karlberg J,
Kristom B, Marcus C, Moell C, Ritzen M, Tuvemo T,
Wattsgard C, Westgren U, Westphal O, Aman J. Improved
®nal height in girls with Turner's syndrome treated with
growth hormone and oxandrolone. J Clin Endocrinol Metab
1996; 81: 635 ± 640.
4 Rosenfeld RG, Attie KM, Frane J, Brasel JA, Burstein S, Casa
JF, Chernausek S, Gotlin RW, Kuntze J, Lippe BM, Mahoney
CP, Moore WV, Saenger P, Johanson AJ. Growth hormone
therapy of Turner's syndrome: bene®cial effect on adult
height. J Pediatr 1998; 132: 319 ± 324.
5 Hall JG, Sybert VP, Williamson RA, Fisher NL, Reed SD.
Turner syndrome. West J Med 1982; 137: 32 ± 44.
6 Hanaki K, Ohzeki T, Ishitani N, Mozozumi H, MatsudaOhtahara H, Shiraki K. Fat distribution in overweight patients
with Ullrich ± Turner syndrome. Am J Med Genet 1992; 42:
428 ± 430.
7 Gravholt CH, Juul S, Naeraa RW, Hansen J. Morbidity in
Turner Syndrome. J Clin Epidemiol 1998; 51: 147 ± 158.
8 Rolland-Cachera MF, Sempe M, Guilloud-Bataille M, Patois
E, Pequinot-Guggenbuhl F. Adiposity indices in children. Am
J Clin Nutr 1982; 36: 178 ± 184.
9 Poskitt EM. De®ning childhood obesity: the relative body
mass index (BMI). European Childhood Obesity Group. Acta
Paediatr 1995; 84: 961 ± 963.
10 August GP, Lippe BM, Blethen SL, Rosenfeld RG, Seelig SA,
Johanson AJ, Compton PG, Frane JW, McClellan BH, Sherman BM. Growth hormone treatment in the United States:
demographic and diagnostic features of 2331 children.
J Pediatr 1990; 116: 899 ± 903.
11 Hammer LD, Kraemer HC, Wilson DM, Ritter PL, Dornbusch
SM. Standardized percentile curves of body mass index
for children and adolescents. Am J Dis Child 1991; 145:
259 ± 263.
12 Becker R, Chambers JM, Wilks AR. The New S Language: A
programming Environment for Data Analysis and Graphics.
Wadworth & Brooks=Cole: Paci®c Grove, CA, 1988.
13 Cole TJ, Freeman JV, Preece MA. Body mass index reference
curves for the U.K. Arch Dis Child 1995; 73: 25 ± 39.
14 Must A, Dallal GE, Dietz WH. Reference data for obesity:
85th and 95th percentiles of body mass index (wt=ht2) and
triceps skinfold thickness. Am J Clin Nutr 1991; 53: 839 ± 846.
15 Must A, Dallal GE, Dietz WH. Reference data for obesity:
85th and 95th percentiles of body mass index (wt=ht2) Ð a
correction. Am J Clin Nutr 1991; 54: 773.
16 Ogle GD, Allen JR, Humphries IR, Lu PW, Brody JN, Morley
K, Howman-Giles R, Cowell CT. Body composition assessment by dual energy X-ray absorptiometry in subjects aged 4 ±
26 y. Am J Clin Nutr 1995; 61: 746 ± 753.
17 Himes J, Dietz W. Guidelines for overweight in adolescent
preventive services: recommendations from an expert committee. Am J Clin Nutr 1994; 59: 307 ± 316.
18 Guo SS, Roche AF, Chumlea WC, Gardner JD, Siervogel
RM. The predictive value of childhood body mass index
values for overweight at age 35 y. Am J Clin Nutr 1994; 59:
810 ± 819.
19 Van den Broeck J, Wit J-M. Anthropometry and body composition in children. Horm Res 1997; 48 (Suppl 1): 33 ± 42.
20 De Ridder CM, Thijssen JH, Bruning PF, Van den Brande JL,
Zonderland ML, Erich WB. Body fat mass, body fat distribution, and pubertal development; a longitudinal study of physical and hormonal sexual maturation of girls.
J Clin Endocrinol Metab 1992; 75: 442 ± 446.
21 Rico H, Revilla M, Villa LF, Hernandez ER, Alvarez de
Buergo M, Villa M. Body composition in children and
Tanner's stages: a study with dual X-ray absorptiometry.
Metabolism 1993; 42: 967 ± 970.
22 Corel LJA, Van den Broeck J, Rongen-Westerlaken C, Massa
G, Wit JM. Body weight in children with Turner syndrome
treated with growth hormone. Int J Obes 1996; 20: 957 ± 962.
23 Rongen-Westerlaken C, Corel L, van den Broeck J, Massa G,
Karlberg J. Reference values for height, height velocity
and weight in Turner's syndrome. Acta Paediatr 1997; 86:
937 ± 942.
24 Sklar CA, Kaplan SL, Grumbach MM. Evidence for dissociation between adrenarche and gonadarche; studies in patients
with idiopathic precocious puberty, gonadal dysgenesis, isolated gonadotrophin de®ciency, and constitutionally delayed
growth and adolescence. J Clin Endocrinl Metab 1980; 51:
548 ± 556.
25 Frisch R, McArthur J. Menstrual cycles: fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science 1974; 85: 949 ± 951.
26 Park E. Body shape in Turner syndrome. Human Biol 1977;
49: 215 ± 223.
235
International Journal of Obesity